ARM: shmobile: "i2c<m>" vs. "iic<n>" DT nodes and pinmux

ARM: shmobile: "i2c<m>" vs. "iic<n>" DT nodes and pinmux

[Geert Uytterhoeven]

Hi,

On r8a7790, DT device nodes and C/DT pinmux data for IIC<n> are called
"iic<n>", with DT aliases from "i2c<m>" to the "iic<n>" DT nodes.
On r8a7791, DT device nodes and C/DT pinmux data for IIC<n> are called
"i2c<m>", with DT aliases from "i2c<m>" to the "i2c<m>" DT nodes.

In light of the proliferation of other members of the R-Car Gen 2 families, is
there a plan to make this consistent?

Thanks!

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: ARM: shmobile: "i2c<m>" vs. "iic<n>" DT nodes and pinmux

[Wolfram Sang]

[-- Attachment #1: Type: text/plain, Size: 729 bytes --]

On Tue, Aug 12, 2014 at 03:11:18PM +0200, Geert Uytterhoeven wrote:
> Hi,
> 
> On r8a7790, DT device nodes and C/DT pinmux data for IIC<n> are called
> "iic<n>", with DT aliases from "i2c<m>" to the "iic<n>" DT nodes.
> On r8a7791, DT device nodes and C/DT pinmux data for IIC<n> are called
> "i2c<m>", with DT aliases from "i2c<m>" to the "i2c<m>" DT nodes.
> 
> In light of the proliferation of other members of the R-Car Gen 2 families, is
> there a plan to make this consistent?

The 'i2c'-prefix describes the IP core handled by the rcar driver. This
core is named i2c in the datasheets.

The 'iic'-prefix describes the IP core handled by the sh-mobile-driver.
This core is named iic(b) in the datasheets.


[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

Re: ARM: shmobile: "i2c<m>" vs. "iic<n>" DT nodes and pinmux

[Wolfram Sang]

[-- Attachment #1: Type: text/plain, Size: 731 bytes --]


On Tue, Aug 12, 2014 at 03:11:18PM +0200, Geert Uytterhoeven wrote:
> Hi,
> 
> On r8a7790, DT device nodes and C/DT pinmux data for IIC<n> are called
> "iic<n>", with DT aliases from "i2c<m>" to the "iic<n>" DT nodes.
> On r8a7791, DT device nodes and C/DT pinmux data for IIC<n> are called
> "i2c<m>", with DT aliases from "i2c<m>" to the "i2c<m>" DT nodes.
> 
> In light of the proliferation of other members of the R-Car Gen 2 families, is
> there a plan to make this consistent?

The 'i2c'-prefix describes the IP core handled by the rcar driver. This
core is named i2c in the datasheets.

The 'iic'-prefix describes the IP core handled by the sh-mobile-driver.
This core is named iic(b) in the datasheets.


[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

Re: ARM: shmobile: "i2c<m>" vs. "iic<n>" DT nodes and pinmux

[Geert Uytterhoeven]

Hi Wolfram,

On Tue, Aug 12, 2014 at 11:47 PM, Wolfram Sang <wsa@the-dreams.de> wrote:
> On Tue, Aug 12, 2014 at 03:11:18PM +0200, Geert Uytterhoeven wrote:
>> On r8a7790, DT device nodes and C/DT pinmux data for IIC<n> are called
>> "iic<n>", with DT aliases from "i2c<m>" to the "iic<n>" DT nodes.
>> On r8a7791, DT device nodes and C/DT pinmux data for IIC<n> are called
>> "i2c<m>", with DT aliases from "i2c<m>" to the "i2c<m>" DT nodes.
>>
>> In light of the proliferation of other members of the R-Car Gen 2 families, is
>> there a plan to make this consistent?
>
> The 'i2c'-prefix describes the IP core handled by the rcar driver. This
> core is named i2c in the datasheets.
>
> The 'iic'-prefix describes the IP core handled by the sh-mobile-driver.
> This core is named iic(b) in the datasheets.

I know the difference between i2c and iic(b).

However, the DT device nodes and C/DT pinmux data for iic(b) on r8a7791
are called i2c, too, instead of iic.

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: ARM: shmobile: "i2c<m>" vs. "iic<n>" DT nodes and pinmux

[Wolfram Sang]

[-- Attachment #1: Type: text/plain, Size: 1670 bytes --]

On Wed, Aug 13, 2014 at 09:19:43AM +0200, Geert Uytterhoeven wrote:
> Hi Wolfram,
> 
> On Tue, Aug 12, 2014 at 11:47 PM, Wolfram Sang <wsa@the-dreams.de> wrote:
> > On Tue, Aug 12, 2014 at 03:11:18PM +0200, Geert Uytterhoeven wrote:
> >> On r8a7790, DT device nodes and C/DT pinmux data for IIC<n> are called
> >> "iic<n>", with DT aliases from "i2c<m>" to the "iic<n>" DT nodes.
> >> On r8a7791, DT device nodes and C/DT pinmux data for IIC<n> are called
> >> "i2c<m>", with DT aliases from "i2c<m>" to the "i2c<m>" DT nodes.
> >>
> >> In light of the proliferation of other members of the R-Car Gen 2 families, is
> >> there a plan to make this consistent?
> >
> > The 'i2c'-prefix describes the IP core handled by the rcar driver. This
> > core is named i2c in the datasheets.
> >
> > The 'iic'-prefix describes the IP core handled by the sh-mobile-driver.
> > This core is named iic(b) in the datasheets.
> 
> I know the difference between i2c and iic(b).
> 
> However, the DT device nodes and C/DT pinmux data for iic(b) on r8a7791
> are called i2c, too, instead of iic.

Ah, yes, now I remember. They are called like this in the datasheet. If
we would stick to the IIC ones, it becomes even more confusing, because:

IIC3 is I2C6
IIC0 is I2C7
IIC1 is I2C8

and there is no IIC2. Plus, in the Koelsch board schematics, the DVFS
bus is also named I2C6, not IIC3. So, I went for that.

On Lager, I also wanted to go for i2c4-7, yet I was convinced to use iic
there. I forgot about this inconsistency. And I have to admit, if we
look at it per-SoC, I think the current solutions make most sense. They
are just not consistent :(


[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

Re: ARM: shmobile: "i2c<m>" vs. "iic<n>" DT nodes and pinmux

[Sergei Shtylyov]

Hello.

On 08/12/2014 05:11 PM, Geert Uytterhoeven wrote:

> On r8a7790, DT device nodes and C/DT pinmux data for IIC<n> are called
> "iic<n>", with DT aliases from "i2c<m>" to the "iic<n>" DT nodes.
> On r8a7791, DT device nodes and C/DT pinmux data for IIC<n> are called
> "i2c<m>", with DT aliases from "i2c<m>" to the "i2c<m>" DT nodes.

> In light of the proliferation of other members of the R-Car Gen 2 families, is
> there a plan to make this consistent?

    I can probably look at this since the SH-Mobile driver (and/or hardware?) 
is clearly of better quality than R-Car one: I'm constantly getting -ENXIO and 
-EBUSY from the latter trying to read images from ADV7180 I2C video decoder.

WBR, Sergei

Re: ARM: shmobile: "i2c<m>" vs. "iic<n>" DT nodes and pinmux

[Sergei Shtylyov]

Hello.

On 08/21/2014 11:09 PM, Sergei Shtylyov wrote:

>> "iic<n>", with DT aliases from "i2c<m>" to the "iic<n>" DT nodes.
>> On r8a7791, DT device nodes and C/DT pinmux data for IIC<n> are called
>> "i2c<m>", with DT aliases from "i2c<m>" to the "i2c<m>" DT nodes.

>> In light of the proliferation of other members of the R-Car Gen 2 families, is
>> there a plan to make this consistent?

>     I can probably look at this since the SH-Mobile driver (and/or hardware?)
> is clearly of better quality than R-Car one: I'm constantly getting -ENXIO and
> -EBUSY from the latter trying to read images from ADV7180 I2C video decoder.

    Indeed, on Lager reading from the same chip seems to work flawlessy. 
However, I was too optimistic about R8A7791: it doesn't have IIC controller 
matching I2C2. :-/
    So the only thing I can do is to fix inconsistent labels in 
R8A7791/Koelsch DTs if Wolfram won't beat me to it.

WBR, Sergei